The hypothesis of this application is that infection of the human gastrointestinal tract by JC virus (JCV) leads to a chronic, latent infection in the colon. Later in life, reactivation of the virus occurs due to a rearrangement in the transcription control region (TCR) of the virus, which leads to expression of the JCV T-antigen in some colonic epithelial cells. Initially, the T-antigen stabilizes nuclear beta-catenin, which permits unregulated proliferation, without loss of the APC gene. T-antigen has other properties that trigger chromosomal instability (CIN), which is the hallmark of aneuploid tumors. In the setting of proliferation and CIN, loss of heterozygosity ('LOH') events occurs at critical tumor suppressor genes, including APC and p53. This application suggests that JCV is the initial cause of the genomic instability that initiates multi-step carcinogenesis in the colon. We have data that JCV DNA is present in 89% of colon cancers, and in the normal colonic tissues of most people. We have found that there are rearrangements in the TCRs of JCV isolates from colon cancers that are not present in TCRs from the normal colon. These TCRs are more transcriptionally active in vitro. We have developed a novel in vitro model using a fetal colonic cell line in which infection by JCV leads to the induction of CIN. We have also induced CIN in a diploid colonic cell line by transfection of the cloned JCV T-antigen. In this application, we propose to rigorously test the hypothesis that the expression of JCV T-antigen correlates with the stabilization of beta-catenin, the later loss of APC, and the initiation of CIN by studying resected specimens of colorectal neoplasia. Second, we propose to determine whether the integration of the virus into the human genome is necessary for expression of T-antigen and the induction of CIN, using both surgically resected tissues and in vitro approaches. Third, we propose to use our in vitro models of JCV infection to study the time course of immortalization and transformation. Finally, we propose to test the hypothesis that rearrangements in the TCR of JCV provide a mechanism whereby a latent infection is converted into an active one, leading to expression of the viral genes. The implication of this work is that JCV, which is commonly present in the gastrointestinal tract of the majority of healthy humans, may be involved in the initiation of CIN, and colon cancer. If this virus plays a mechanistic role in human carcinogenesis, this could lead to novel preventive and treatment strategies.